Abstract
OBJECTIVE: Recognizing smoking as a risk factor for rheumatoid arthritis (RA) severity, the present study was undertaken to evaluate patient- and health care-level predictors of smoking cessation in patients with RA to guide implementation of smoking cessation interventions. METHODS: Electronic health record data from 2 health systems were abstracted for patients with at least 2 International Classification of Disease diagnosis codes for RA between 2005 and 2016. Patients missing smoking statuses or with <6 months of follow-up were excluded. Multivariable logistic regression was used to determine predictors of smoking cessation. RESULTS: Among 3,577 patients with RA, 507 smoked at baseline, and 29% quit over a median of 4.75 years. Black male patients, ages 40-59 years and enrolled in Medicaid, were significantly more likely to be baseline smokers; however, none of these factors predicted cessation. Instead, patients new to rheumatology care were 60% more likely to quit (adjusted odds ratio [OR(adj) ] 1.60 [95% confidence interval (95% CI) 1.02-2.50]), and patients in the rural community health system were 66% more likely to quit (OR(adj) 1.66 [95% CI 1.03-2.69]). Seropositive patients were 43% less likely to quit smoking (OR(adj) 0.57 [95% CI 0.35-0.91]). CONCLUSION: Health care factors, including health system and being new to rheumatology care, were more predictive of smoking cessation in patients with RA than patient sociodemographic factors, suggesting an important role for health system cessation efforts for patients with RA. Seropositive patients were less likely to quit and may particularly benefit from cessation support. Emphasizing smoking cessation with new or seropositive RA patients and leveraging health system interventions could improve smoking cessation and outcomes in RA.